1. INTRODUCTION TO CLOUD COMPUTING Cs 595 Lecture 5 2/11/2015

2. PAAS  Platform as a Service (PaaS) aka cloudware  Give me nice API and take care of the implementation  Supplies all resourced needed to build apps and services without having to download or install software  Provides a computing platform and solution stack  E.g for web application need OS, web server, DB, prog language  Provides support to create user interfaces (HTML, Javascript)  Provides automatic facilities for concurrency management, scalability, failover, and security  Services include:  app design, development, testing, deployment, hosting

3. PAAS  Customer interacts with platform through API  Runtime services – allows application to leverage infrastructure  Platform manages and scales  Team collaboration, web service integration, database integration, security, scalability, storage, state management, versioning

4. PAAS  Supports web development interfaces  SOAP (simple object access protocol), REST (Representational state transfer), allow construction of multiple web services (mashups)  Interfaces able to access DBs, reuse services

5. PAAS PROVIDES  Development teams across world to work together  Merge web services from multiple sources  Cost savings from using built-in security, scalability and failover  Cost-savings from using higher-level programming abstractions

6. PROBLEMS WITH PAAS  Vendors used proprietary services or languages – developer may be locked in  Lack of portability and interoperability – if develop on one cloud, can’t move to another (unless pay …) – Lock-in  What if provider goes out of business?

7.  Examples:  Google App Engine  Heroku  RightScale  Salesforce.com

8. SAAS  Software as a Service (SaaS) – web based applications  Just run it for me!  Software available on cloud for use  Application hosted as a service to customers who access via the internet  Single instance runs and services multiple end users

9. SAAS  Good candidates for SaaS:  Simple task with little interaction with other systems  Customers who want high powered apps but do not want to develop  Customer resource management CRM  Video conferencing  IT service management  Accounting  Web analytics  Web content management

10. SAAS  Unlike earlier distributed computing tools, SaaS specifically uses web/Internet tools  Built with multitenant in mind  Can access from anywhere as long as have access to Internet  SaaS often used as a component of another application – mashup or plugin

11. BENEFITS TO SAAS  Everyone knows WWW, little training needed  Smaller IT staff needed  Easier to customize  Better marketing by providers, accommodate more  Web reliability  Security (SSL used), don’t need VPNs (Virtual private networks on back-end)

12. SAAS  Pros/Cons  Customer doesn’t have to maintain or support SW  Out of customer’s hands when hosting service changes it  Use software out of box  Instead of just paying for its once, billed  Don’t have to pay as much up front, cheaper more reliable

13. OBSTACLES TO SAAS  Specific computational need not addressed  Lock-in – can’t move to new vendor without penalty  Open source and cheaper hardware

14. EXAMPLE APPLICATIONS BENEFITING  Using Hadoop tool, open-source MapReduce  NY Times converted 11 M articles, images in archive to PDF  Instead of 7 weeks, using Hadoop took 24 hours, < $300  Animoto’s mashup tool – create videos from set of images and music  Scaled from 50 to 3500 servers in 3 days  Application built to be horizontal

15.  Examples:  Gmail  Dropbox  Microsoft Office 365

16. FUTURE OF SAAS  Move all processing power to the cloud and carry ultralight input device  Already happening?  E-mail  Google Docs  OnLive  Implications for Microsoft, software as purchasable local application  Windows Live (Microsoft’s cloud)  Adobe web based photoshop

17. IAAS, PAAS, SAAS

18. IN SUMMARY - IAAS, PAAS, SAAS  With IaaS  Provider doesn’t know what you are going to do with HW  Just ask for resources, including OS (VMs)  So you can specify how many machines, how many VMs per machine, etc.  Can create your own PaaS, or SaaS on IaaS

19. IAAS, PAAS, SAAS  With PaaS  Ask for specific web services, DBs, etc.  Restricted to using only those, can modify only within constraints of platform  System decides what hardware and how many VMs you get, e.g. scaling  With SaaS  Just say which software and you use it

21. PAPER TO READ  A Survey of Cloud Computing Architectures